Microwave ablation versus sorafenib for intermediate-Stage Hepatocellular carcinoma with transcatheter arterial chemoembolization refractoriness: a propensity score matching analysis

Figures & data

Figure 1. Patient selection flowchart. BCLC: Barcelona Clinic Liver Cancer; TACE: transarterial chemoembolization; MWA: microwave ablation; HAIC: hepatic artery infusion chemotherapy.

Figure 2. Images in a 64 year-old male patient with BCLC stage B hepatocellular carcinoma (HCC) before TACE, who has two lesions measuring approximately 6.1 cm and 1.0 cm (no shown), show microwave ablation (MWA) as first-line treatment after transarterial chemoembolization (TACE) refractoriness. (A) Arterial-phase CT image obtained about one month after first TACE shows that necrotic volume is less than 20% of large lesion, and small lesion disappears (no shown). (B) He underwent CT-guided multiple-position MWA for non-necrotic areas of large lesion with the output of 60 W for 69 min in total. (C) Arterial-phase CT image obtained about 1 month after MWA shows that complete ablation was achieved. (D) Arterial-phase MRI image obtained about 34 months after MWA (5 December 2019) shows no tumor recurrence.

Table 1. Baseline Characteristics of patients before and after Propensity Score Matching (PSM).

	Before PSM			After PSM
Characteristics	Sorafenib (n = 22)	MWA (n = 30)	p	Sorafenib (n = 22)	MWA (n = 22)	p
Age			.189			.322
<50 years	8 (36.4)	6 (20.0)		8 (36.4)	5 (22.7)
≥50 years	14 (63.6)	24 (80.0)		14 (63.6)	17 (77.3)
Gender			1.000			1.000
Female	3 (13.6)	4 (13.3)		3 (13.6)	3 (13.6)
Male	19 (86.4)	26 (86.7)		19 (86.4)	19 (86.4)
Hepatitis B virus			.629			1.000
Positive	21 (95.5)	27 (90.0)		21 (95.5)	20 (90.9)
Negative	1 (4.5)	3 (10.0)		1 (4.5)	2 (9.1)
Child-Pugh class			1.000			1.000
A	21 (95.5)	28 (93.3)		21 (95.5)	20 (90.9)
B	1 (4.5)	2 (6.7)		1 (4.5)	2 (9.1)
Number			.173			.545
≤2	9 (40.9)	18 (60.0)		9 (40.9)	11 (50.0)
>2	13 (59.1)	12 (40.0)		13 (59.1)	11 (50.0)
Tumor Size			.756			1.000
≤3cm	9 (40.9)	11 (36.7)		9 (40.9)	9 (40.9)
>3cm	13 (59.1)	19 (63.3)		13 (59.1)	13 (59.1)
a-fetoprotein level			.539			.472
<400ng/mL	16 (72.7)	24 (80.0)		16 (72.7)	18 (81.8)
≥400ng/mL	6 (27.3)	6 (20.0)		6 (27.3)	4 (18.2)

Values in parentheses are percentages. Parameter data were compared by t test. Categorical data were compared by using the ${\mathrm{\chi }}^2$ test or the Fisher exact test, as appropriate.

Figure 3. Kaplan-Meier curves of overall-all survival (OS) and progression-free survival (PFS) in patents who took Sorafenib or accepted microwave ablation (MWA) before one-to-one PSM (Sorafenib-group: n = 22; the median OS, 16.6 months; 95% CI: 13.4–19.8 months). MWA-group: n = 30; the median OS, 48.8 months; 95% CI: not available. p = .001) (A) and (Sorafenib-group: n = 22; the median PFS, 2.8 months; 95% CI: 1.2–4.4 months. MWA group: n = 30; the median PFS, 9.3 months; 95% CI: 0.0–27.2 months. p = .001) (C), respectively. After one-to-one PSM, Kaplan-Meier curves of overall-all survival (OS) and progression-free survival (PFS) in patents who took Sorafenib or accepted microwave ablation (MWA) (Sorafenib-group: n = 22; the median OS, 16.6 months; 95% CI: 13.4–19.8 months. MWA-group: n = 22; the median OS, Not reached; 95% CI: not available. p = .001) (B) and (Sorafenib group: n = 22; the median PFS, 2.8 months; 95% CI: 1.2–4.4 months. MWA group: n = 22; the median PFS, 9.0 months; 95% CI: 6.6–11.4 months. p = .006) (D), respectively.

Table 2. Univariate and Multivariate analysis of OS in the enrolled cohort.

Variable	No. of Cases	Univariate Analysis		Multivariate Analysis
		HR (95% CI)	p Value	HR (95% CI)	p Value
Gender (Male vs. Female)	45 vs. 7	0.96 (0.29–3.19)	0.944	–	–
Age (<50 vs. ≥50)	14 vs. 38	2.76 (1.25–6.09)	0.012	2.93 (1.28–6.73)	0.011
Child Pugh Class (A vs. B)	49 vs. 3	0.81 (0.11–6.06)	0.835	–	–
Tumor Size (≤3cm vs. >3cm)	20 vs. 32	0.97 (0.46–2.05)	0.933	–	–
AFP level (<400ng/ml vs. ≥400ng/ml)	40 vs. 12	0.45 (0.19–1.08)	0.072	–	0.075
HBV (Absent vs. Present)	4 vs. 48	1.46 (0.44–4.85)	0.541	–	–
Numbers (≤2 vs. >2)	27 vs. 25	1.09 (0.52–2.28)	0.825	–	–
Treatment (MWA vs. Sorafenib)	30 vs. 22	0.29 (0.14–0.64)	0.002	0.28 (0.13–0.62)	0.002

HR: hazard ratio; AFP: a-fetoprotein; MWA: Microwave Ablation. Numbers in parentheses were 95% CIs. The data were analyzed using the Cox proportional hazard model, and the p values ≤0.2 of the univariate predictors were entered into the multivariate model. Significance level: 0.05.

Table 3. Univariate and Multivariate analysis of PFS in the enrolled cohort.

Variable	No. of Cases	Univariate Analysis		Multivariate Analysis
		HR (95% CI)	P value	HR (95% CI)	P value
Gender (Male vs. Female)	45 vs. 7	1.25 (0.44–3.55)	0.674	–	–
Age (<50 vs. ≥50)	14 vs. 38	1.93 (1.00–3.71)	0.050	2.05 (1.05–4.01)	0.035
Child Pugh Class (A vs. B)	49 vs. 3	0.98 (0.23–4.10)	0.974	–	–
Tumor Size (≤3cm vs. >3cm)	20 vs. 32	1.03 (0.56–1.90)	0.923	–	–
AFP level (<400ng/ml vs. ≥400ng/ml)	40 vs. 12	0.76 (0.36–1.59)	0.460	–	–
HBV (Absent vs. Present)	4 vs. 48	1.41 (0.50–3.98)	0.517	–	–
Numbers (≤2 vs. >2)	27 vs. 25	0.98 (0.53–1.79)	0.941	–	–
Treatment (MWA vs. Sorafenib)	30 vs. 22	0.36 (0.19–0.68)	0.002	0.35 (0.18–0.66)	0.001

HR: hazard ratio; AFP: a-fetoprotein; MWA: Microwave Ablation. Numbers in parentheses are 95% CIs. The data were analyzed using the Cox proportional hazard model, and the p values ≤0.2 of the univariate predictors were entered into the multivariate model. Significance level: 0.05.

Data availability

The datasets used and analyzed in the current study are available from the corresponding author on reasonable request.